Strange heating problem

[TRK's dad]

Since replacing the tapstat for the hot water cylinder (see previous thread) I have encountered a rather bizarre problem. When the hot water / heating is switched off (either by the timer or manually) water flows back through the circuit into the feed & expansion tank in the loft. This happens quite quickly, last night I reduced the level in the feed & expansion tank to about one third capacity and within 90 minutes of the system switching off the overflow was running again. Switch the system back on again and all is well, switch it off and you can watch the water coming back up the circuit into the tank. So, we currently have the system running permanently with the rads turned off through TRVs. Does anyone have any ideas about what's causing this ?

The ball valve into the feed & expansion tank is shutting off the supply from the mains at an appropriate level.

Further relevant information could be that before replacing the tapstat the boiler thermostat was set very hot (about 170 degrees, the boiler is of sufficient age for that to be Farenheit) by the previous owners of the house and we hadn't changed it. We've turned it down since replacing the tapstat as the plumber who did it thought that a bit hot. Were the previous owners of the house running the boiler at an excessive temperature to avoid this overflow problem I wonder ? If so, how does that work ?

Thanks for any help, I'm rather baffled.

[Set Square]

What are the relative levels of the water in your two header tanks? Is the main DHW feeder level *higher* than that in the primary circuit F&E tank?

If the main tank is higher, this sounds like a case of a leaking coil inside your hot water cylinder, allowing water to flow from your slightly higher pressure hot water system into your primary circuit. The fact that it doesn't do it when the pump is running is probably because the pump introduces a slightly higher pressure into that part of the primary circuit - and stops, or maybe reverses, the flow.

In order to test whether the problem *is* caused by a leaking coil, temporarily bung up the outlet of the main cold water tank with a cork. [You won't be able to use any hot water whilst the cork is in place!] If this stops it happening, even with the pump off, it makes the diagnosis fairly certain. The only cure would then be to replace the hot water cylinder.

Nor quite sure what you mean by "tapstat". Are you referring to the thermostat which is strapped to the hot water cylinder? If so, I can't see how changing that could have caused this problem.

[Dave Liquorice]

This was my first thoughts as well.

There was a mechanical valve, like a radiator thermostatic valve with remote sensor, in the primary. This failed and was replaced(?). I suspect the banging about of pipe work to change this valve has caused a failure in the cylinder. It all seems a very old system, tens of years, so disturbing things may well cause 'em to fail.

[TRK's dad]

Thanks for your reply.

You're right about what I referred to as a tapstat - see here

don't necessarily think that replacing it has caused the problem, it's merely come to light as a result of fiddling about with the system while / after its replacement.

The level of the water in the main tank *is* higher than that in the F&E tank (they're both resting on the same wooden plinth in the loft but by virtue of the fact that the main tank is much bigger and taller the level of the water in it is higher than the F&E tank). I had read that this may be caused by a leaking coil, however the water that goes round the coil (& ends up in the F&E tank) is so filthy that if any of it was leaking into the hot water cylinder the contamination would be easy to spot when turning on a hot tap I should think. Would it be worth putting some food dye in the F&E tank to see if coloured water then came out of the hot taps ?

I will try your cork method tomorrow (oh no, that means opening a bottle of wine tonight!) to see what happens.

I am not adverse to replacing the tank, the current one is extremely elderly but I plan on replacing the lot in spring next year so I'm just trying to nurse it along at the moment.

Thanks

[Set Square]

But it appears that this is a plumbing component, installed in the primary circuit - rather than just being an electrical device. That being the case, the tank connections - and hence the internal coil - may well have been strained during the fitting operation, and could have caused the leak.

Yes, but the leakage is the other way! *Clean* water is leaking from the hot cylinder *into* the coil containing the dirty water - and thence up to the F&E tank. When the pump is on, it's probably providing enough pressure to balance the pressure differential, and stop the flow, but not enough to make it flow the other way.

I think this will probably be conclusive - as long as you're not too drunk to observe the results!

If your tank *is* leaking, you really *must* replace it without waiting. Otherwise, fresh oxygenated water will be entering your primary circuit all the time - and you will have much greater expense due to corroded boiler and radiators.

[TRK's dad]

[snip]

I conducted the above experiment, although my cold water tank has three outlets for some reason and I managed to close of three quite well but I don't think that the third was entirely sealed.

Upon turning off the pump water flowed from the primary circuit back into the feed and expansion tank, although not as quickly as before. Also, as soon as the pump was turned off a succession of air bubbles came back up the circuit into the F&E tank.

Does this mean that the coil in the hot water cylinder is broken? I guess it does.

[Set Square]

I don't think we *quite* have the conclusive evidence yet.

You need to find which of the 3 outlets from the cold header is the cold feed to the base of the hot cylinder. It is *this* one which you must be able to seal off very well. The others possibly feed shower pumps or cold taps, and don't matter.

What you are looking for is this:

• When the cold feed to the hot cylinder is open, and the CH pump is not running, the water level in the small F&E tank rises and overflows

• When the cold feed to the hot cylinder is sealed off, and the pump is not running, the water in the small F&E tank does *not* rise and overflow

If these conditions are met, it would strongly suggest a leaking coil in the hot water cylinder.

[TRK's dad]

It should be quite easy to discover which one of the three is for the cold taps - block each one up in succession and run a cold tap 'til no water comes out. The shower one I'm not sure - I could block one up and turn the shower control down to the coldest setting and see if it stays hot. Once I've discovered those two then by process of elimination I've got the cold feed for the cylinder. As I recall all three outlets are the same size so no clues there. A couple of them do have shutoff valves on them - does that offer any clues as to their purpose ?

The reason I'm perservering on my own is that my in laws are about to have solar panels for hot water fitted which will involve a new cylinder. They only replaced their conventional cylinder about 6 months ago so it would be good if I can get hold of that, it should tide me over until spring (as long as it fits my system). As my father in law can do plumbing and fitted his existing hot water cylinder himself I might be able to talk him into fitting their existing one at my house as well !

Thanks for your perserverance as well.

[Set Square]

The one you want is for the *hot* taps. [The hot taps only run when cold water from the header tank flows into the base of the hot cylinder, expelling hot water from the top to the taps]. Whichever outlet - when blocked up - stops the flow of water from the hot taps is the one you need to block up to test for coil leak. If you follow this pipe down from the header, you will find that it connects into the side of the hot cylinder, right at the bottom.